[codex] Improve agent runtime recovery and governance (#4086)

## Thinking Path

> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat runtime, agent import path, and agent configuration
defaults determine whether work is dispatched safely and predictably.
> - Several accumulated fixes all touched agent execution recovery, wake
routing, import behavior, and runtime concurrency defaults.
> - Those changes need to land together so the heartbeat service and
agent creation defaults stay internally consistent.
> - This pull request groups the runtime/governance changes from the
split branch into one standalone branch.
> - The benefit is safer recovery for stranded runs, bounded high-volume
reads, imported-agent approval correctness, skill-template support, and
a clearer default concurrency policy.

## What Changed

- Fixed stranded continuation recovery so successful automatic retries
are requeued instead of incorrectly blocking the issue.
- Bounded high-volume issue/log reads across issue, heartbeat, agent,
project, and workspace paths.
- Fixed imported-agent approval and instruction-path permission
handling.
- Quarantined seeded worktree execution state during worktree
provisioning.
- Queued approval follow-up wakes and hardened SQL_ASCII heartbeat
output handling.
- Added reusable agent instruction templates for hiring flows.
- Set the default max concurrent agent runs to five and updated related
UI/tests/docs.

## Verification

- `pnpm install --frozen-lockfile`
- `pnpm exec vitest run server/src/__tests__/company-portability.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/heartbeat-comment-wake-batching.test.ts
server/src/__tests__/heartbeat-list.test.ts
server/src/__tests__/issues-service.test.ts
server/src/__tests__/agent-permissions-routes.test.ts
packages/adapter-utils/src/server-utils.test.ts
ui/src/lib/new-agent-runtime-config.test.ts`
- Split integration check: merged this branch first, followed by the
other [PAP-1614](/PAP/issues/PAP-1614) branches, with no merge
conflicts.
- Confirmed this branch does not include `pnpm-lock.yaml`.

## Risks

- Medium risk: touches heartbeat recovery, queueing, and issue list
bounds in central runtime paths.
- Imported-agent and concurrency default behavior changes may affect
existing automation that assumes one-at-a-time default runs.
- No database migrations are included.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- OpenAI Codex, GPT-5.4 tool-enabled coding model, agentic
code-editing/runtime with local shell and GitHub CLI access; exact
context window and reasoning mode are not exposed by the Paperclip
harness.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge

---------

Co-authored-by: Paperclip <noreply@paperclip.ing>
This commit is contained in:
Dotta 2026-04-20 06:19:48 -05:00 committed by GitHub
parent 057fee4836
commit 16b2b84d84
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
38 changed files with 1569 additions and 240 deletions

View file

@ -38,6 +38,9 @@ import { getDefaultCompanyGoal } from "./goals.js";
const ALL_ISSUE_STATUSES = ["backlog", "todo", "in_progress", "in_review", "blocked", "done", "cancelled"];
const MAX_ISSUE_COMMENT_PAGE_LIMIT = 500;
export const ISSUE_LIST_DEFAULT_LIMIT = 500;
export const ISSUE_LIST_MAX_LIMIT = 1000;
const ISSUE_LIST_RELATED_QUERY_CHUNK_SIZE = 500;
export const MAX_CHILD_ISSUES_CREATED_BY_HELPER = 25;
const MAX_CHILD_COMPLETION_SUMMARIES = 20;
const CHILD_COMPLETION_SUMMARY_BODY_MAX_CHARS = 500;
@ -106,6 +109,10 @@ type IssueUserCommentStats = {
myLastCommentAt: Date | null;
lastExternalCommentAt: Date | null;
};
type IssueReadStat = {
issueId: string;
myLastReadAt: Date | null;
};
type IssueLastActivityStat = {
issueId: string;
latestCommentAt: Date | null;
@ -158,6 +165,18 @@ function escapeLikePattern(value: string): string {
return value.replace(/[\\%_]/g, "\\$&");
}
export function clampIssueListLimit(limit: number): number {
return Math.min(ISSUE_LIST_MAX_LIMIT, Math.max(1, Math.floor(limit)));
}
function chunkList<T>(values: T[], size: number): T[][] {
const chunks: T[][] = [];
for (let index = 0; index < values.length; index += size) {
chunks.push(values.slice(index, index + size));
}
return chunks;
}
function truncateInlineSummary(value: string | null | undefined, maxChars = CHILD_COMPLETION_SUMMARY_BODY_MAX_CHARS) {
const normalized = value?.trim();
if (!normalized) return null;
@ -494,20 +513,22 @@ function latestIssueActivityAt(...values: Array<Date | string | null | undefined
async function labelMapForIssues(dbOrTx: any, issueIds: string[]): Promise<Map<string, IssueLabelRow[]>> {
const map = new Map<string, IssueLabelRow[]>();
if (issueIds.length === 0) return map;
const rows = await dbOrTx
.select({
issueId: issueLabels.issueId,
label: labels,
})
.from(issueLabels)
.innerJoin(labels, eq(issueLabels.labelId, labels.id))
.where(inArray(issueLabels.issueId, issueIds))
.orderBy(asc(labels.name), asc(labels.id));
for (const issueIdChunk of chunkList(issueIds, ISSUE_LIST_RELATED_QUERY_CHUNK_SIZE)) {
const rows = await dbOrTx
.select({
issueId: issueLabels.issueId,
label: labels,
})
.from(issueLabels)
.innerJoin(labels, eq(issueLabels.labelId, labels.id))
.where(inArray(issueLabels.issueId, issueIdChunk))
.orderBy(asc(labels.name), asc(labels.id));
for (const row of rows) {
const existing = map.get(row.issueId);
if (existing) existing.push(row.label);
else map.set(row.issueId, [row.label]);
for (const row of rows) {
const existing = map.get(row.issueId);
if (existing) existing.push(row.label);
else map.set(row.issueId, [row.label]);
}
}
return map;
}
@ -537,27 +558,29 @@ async function activeRunMapForIssues(
.filter((id): id is string => id != null);
if (runIds.length === 0) return map;
const rows = await dbOrTx
.select({
id: heartbeatRuns.id,
status: heartbeatRuns.status,
agentId: heartbeatRuns.agentId,
invocationSource: heartbeatRuns.invocationSource,
triggerDetail: heartbeatRuns.triggerDetail,
startedAt: heartbeatRuns.startedAt,
finishedAt: heartbeatRuns.finishedAt,
createdAt: heartbeatRuns.createdAt,
})
.from(heartbeatRuns)
.where(
and(
inArray(heartbeatRuns.id, runIds),
inArray(heartbeatRuns.status, ACTIVE_RUN_STATUSES),
),
);
for (const runIdChunk of chunkList([...new Set(runIds)], ISSUE_LIST_RELATED_QUERY_CHUNK_SIZE)) {
const rows = await dbOrTx
.select({
id: heartbeatRuns.id,
status: heartbeatRuns.status,
agentId: heartbeatRuns.agentId,
invocationSource: heartbeatRuns.invocationSource,
triggerDetail: heartbeatRuns.triggerDetail,
startedAt: heartbeatRuns.startedAt,
finishedAt: heartbeatRuns.finishedAt,
createdAt: heartbeatRuns.createdAt,
})
.from(heartbeatRuns)
.where(
and(
inArray(heartbeatRuns.id, runIdChunk),
inArray(heartbeatRuns.status, ACTIVE_RUN_STATUSES),
),
);
for (const row of rows) {
map.set(row.id, row);
for (const row of rows) {
map.set(row.id, row);
}
}
return map;
}
@ -617,6 +640,131 @@ function withActiveRuns(
}));
}
async function userCommentStatsForIssues(
dbOrTx: any,
companyId: string,
userId: string,
issueIds: string[],
): Promise<IssueUserCommentStats[]> {
const stats: IssueUserCommentStats[] = [];
for (const issueIdChunk of chunkList(issueIds, ISSUE_LIST_RELATED_QUERY_CHUNK_SIZE)) {
const rows = await dbOrTx
.select({
issueId: issueComments.issueId,
myLastCommentAt: sql<Date | null>`
MAX(CASE WHEN ${issueComments.authorUserId} = ${userId} THEN ${issueComments.createdAt} END)
`,
lastExternalCommentAt: sql<Date | null>`
MAX(
CASE
WHEN ${issueComments.authorUserId} IS NULL OR ${issueComments.authorUserId} <> ${userId}
THEN ${issueComments.createdAt}
END
)
`,
})
.from(issueComments)
.where(
and(
eq(issueComments.companyId, companyId),
inArray(issueComments.issueId, issueIdChunk),
),
)
.groupBy(issueComments.issueId);
stats.push(...rows);
}
return stats;
}
async function userReadStatsForIssues(
dbOrTx: any,
companyId: string,
userId: string,
issueIds: string[],
): Promise<IssueReadStat[]> {
const stats: IssueReadStat[] = [];
for (const issueIdChunk of chunkList(issueIds, ISSUE_LIST_RELATED_QUERY_CHUNK_SIZE)) {
const rows = await dbOrTx
.select({
issueId: issueReadStates.issueId,
myLastReadAt: issueReadStates.lastReadAt,
})
.from(issueReadStates)
.where(
and(
eq(issueReadStates.companyId, companyId),
eq(issueReadStates.userId, userId),
inArray(issueReadStates.issueId, issueIdChunk),
),
);
stats.push(...rows);
}
return stats;
}
async function lastActivityStatsForIssues(
dbOrTx: any,
companyId: string,
issueIds: string[],
): Promise<IssueLastActivityStat[]> {
const byIssueId = new Map<string, IssueLastActivityStat>();
for (const issueIdChunk of chunkList(issueIds, ISSUE_LIST_RELATED_QUERY_CHUNK_SIZE)) {
const [commentRows, logRows] = await Promise.all([
dbOrTx
.select({
issueId: issueComments.issueId,
latestCommentAt: sql<Date | null>`MAX(${issueComments.createdAt})`,
})
.from(issueComments)
.where(
and(
eq(issueComments.companyId, companyId),
inArray(issueComments.issueId, issueIdChunk),
),
)
.groupBy(issueComments.issueId),
dbOrTx
.select({
issueId: activityLog.entityId,
latestLogAt: sql<Date | null>`MAX(${activityLog.createdAt})`,
})
.from(activityLog)
.where(
and(
eq(activityLog.companyId, companyId),
eq(activityLog.entityType, "issue"),
inArray(activityLog.entityId, issueIdChunk),
sql`${activityLog.action} NOT IN (${sql.join(
ISSUE_LOCAL_INBOX_ACTIVITY_ACTIONS.map((action) => sql`${action}`),
sql`, `,
)})`,
),
)
.groupBy(activityLog.entityId),
]);
for (const row of commentRows) {
byIssueId.set(row.issueId, {
issueId: row.issueId,
latestCommentAt: row.latestCommentAt,
latestLogAt: null,
});
}
for (const row of logRows) {
const existing = byIssueId.get(row.issueId);
if (existing) existing.latestLogAt = row.latestLogAt;
else {
byIssueId.set(row.issueId, {
issueId: row.issueId,
latestCommentAt: null,
latestLogAt: row.latestLogAt,
});
}
}
}
return [...byIssueId.values()];
}
export function issueService(db: Db) {
const instanceSettings = instanceSettingsService(db);
@ -1105,99 +1253,12 @@ export function issueService(db: Db) {
const issueIds = withRuns.map((row) => row.id);
const [statsRows, readRows, lastActivityRows] = await Promise.all([
contextUserId
? db
.select({
issueId: issueComments.issueId,
myLastCommentAt: sql<Date | null>`
MAX(CASE WHEN ${issueComments.authorUserId} = ${contextUserId} THEN ${issueComments.createdAt} END)
`,
lastExternalCommentAt: sql<Date | null>`
MAX(
CASE
WHEN ${issueComments.authorUserId} IS NULL OR ${issueComments.authorUserId} <> ${contextUserId}
THEN ${issueComments.createdAt}
END
)
`,
})
.from(issueComments)
.where(
and(
eq(issueComments.companyId, companyId),
inArray(issueComments.issueId, issueIds),
),
)
.groupBy(issueComments.issueId)
? userCommentStatsForIssues(db, companyId, contextUserId, issueIds)
: Promise.resolve([]),
contextUserId
? db
.select({
issueId: issueReadStates.issueId,
myLastReadAt: issueReadStates.lastReadAt,
})
.from(issueReadStates)
.where(
and(
eq(issueReadStates.companyId, companyId),
eq(issueReadStates.userId, contextUserId),
inArray(issueReadStates.issueId, issueIds),
),
)
? userReadStatsForIssues(db, companyId, contextUserId, issueIds)
: Promise.resolve([]),
Promise.all([
db
.select({
issueId: issueComments.issueId,
latestCommentAt: sql<Date | null>`MAX(${issueComments.createdAt})`,
})
.from(issueComments)
.where(
and(
eq(issueComments.companyId, companyId),
inArray(issueComments.issueId, issueIds),
),
)
.groupBy(issueComments.issueId),
db
.select({
issueId: activityLog.entityId,
latestLogAt: sql<Date | null>`MAX(${activityLog.createdAt})`,
})
.from(activityLog)
.where(
and(
eq(activityLog.companyId, companyId),
eq(activityLog.entityType, "issue"),
inArray(activityLog.entityId, issueIds),
sql`${activityLog.action} NOT IN (${sql.join(
ISSUE_LOCAL_INBOX_ACTIVITY_ACTIONS.map((action) => sql`${action}`),
sql`, `,
)})`,
),
)
.groupBy(activityLog.entityId),
]).then(([commentRows, logRows]) => {
const byIssueId = new Map<string, IssueLastActivityStat>();
for (const row of commentRows) {
byIssueId.set(row.issueId, {
issueId: row.issueId,
latestCommentAt: row.latestCommentAt,
latestLogAt: null,
});
}
for (const row of logRows) {
const existing = byIssueId.get(row.issueId);
if (existing) existing.latestLogAt = row.latestLogAt;
else {
byIssueId.set(row.issueId, {
issueId: row.issueId,
latestCommentAt: null,
latestLogAt: row.latestLogAt,
});
}
}
return [...byIssueId.values()];
}),
lastActivityStatsForIssues(db, companyId, issueIds),
]);
const statsByIssueId = new Map(statsRows.map((row) => [row.issueId, row]));
const lastActivityByIssueId = new Map(lastActivityRows.map((row) => [row.issueId, row]));